Introduction

Contemporary business is evolving at an unprecedented pace with the development of global economic, information, and telecommunications technology, which is dramatically changing the nature of work (Griffin et al., 2007). Jobs are becoming more dynamic and ambiguous, entailing roles, tasks, and projects that are constantly shifting (Grant & Parker, 2009). Thus, organizations are increasingly relying on their employees’ initiative to adapt to uncertain and dynamic environments (Grant & Parker, 2009; Griffin et al., 2007). To capture how employees cope with emergent demands and opportunities at work, scholars have introduced the concept of job crafting, i.e., “the actions employees take to shape, mold and redefine their jobs” (Wrzesniewski & Dutton, 2001, p. 180). Job crafting has been shown to promote job satisfaction (Tims et al., 2013a), work engagement (Bakker & Oerlemans, 2019), person–job fit (Lu et al., 2014), and job performance (Bakker et al., 2012).

Increasingly, jobs are becoming inextricably intertwined with interpersonal relationships, connections, and social interactions (Grant & Parker, 2009). This has prompted organizational researchers and practitioners to consider the interpersonal consequences of individual job crafting (e.g., Fong et al., 2021; Tims et al., 2015; cf. Tims & Parker, 2020). Within an interdependent work context, such as teamwork, employees are required to collaborate (Tims & Parker, 2020). However, to protect their individual interests and needs, employees also compete with one another for limited job resources such as challenging assignments, job discretion, social recognition, and career development opportunities. Witnessing others carry out job crafting activities is likely to trigger a social comparison process, in which the focal employee may perceive that the job crafter is in a more desirable and advantageous position to gain resources. This perception may result in negative psychological and behavioral reactions toward the referent coworker.

To examine how employees react to coworkers’ job crafting, we drew on relative deprivation theory (Crosby, 1984), fundamental to which is the tendency of individuals to compare themselves with a referent (e.g., coworkers). When individuals believe that their situation is worse than they deserve, they feel deprived and consequently angry and resentful (Crosby, 1984; Hu et al., 2015; Zoogah, 2010). We propose that focal employees are likely to experience feelings of relative deprivation when their coworkers engage in job crafting. Indeed, job crafting is a way to accumulate resources at work (Tims et al., 2012, 2015), which are limited in the work unit. Such feelings of relative deprivation may lead to negative interpersonal behaviors among focal employees, i.e., fewer prosocial behaviors and more social undermining. Relative deprivation theory also suggests that the extent to which employees feel relative deprivation compared with others depends on their perception of the availability of desirable resources (Buunk et al., 2003; Crosby, 1984). This perception can be examined in terms of their zero-sum mindset, which is defined as the belief that the resource gain of one party is equivalent to another’s loss (Davidai & Ongis, 2019; Sirola & Pitesa, 2017). Employees with a high zero-sum mindset strongly believe that resources are limited and that one person gains resources at the expense of others. Therefore, they are more likely to regard coworkers’ job crafting as depriving them of the resources that they deserve, which may strengthen their feelings of relative deprivation and thus cause more negative interpersonal outcomes. We therefore propose that the effect of coworkers’ job crafting on focal employees’ relative deprivation depends on focal employees’ zero-sum mindset.

Our study makes three contributions to the literature. First, studies examining the interpersonal consequences of job crafting have relied solely on a dyadic level of analysis. That is, they have focused on the reactions of the leader (Fong et al., 2021) or of the colleague with whom the job crafter often interacts at work (Tims et al., 2015). We went beyond these studies by considering job crafting in a broader team context and using multilevel analyses to examine its interpersonal consequences. Specifically, we explored both avoidance and approach job crafting from an interpersonal perspective, thereby responding to the call raised by Tims and Parker (2020), and drew upon relative deprivation theory to advance knowledge of the negative effects of job crafting in a team context. As Frese and Fay (2001, p. 141) argued, initiative among employees “is not always welcomed by supervisors or colleagues,” and high-initiative employees may be “perceived by their environments as being tiring and strenuous.” Job crafting scholars have also suggested that some dimensions of job crafting are more detrimental than others (Tims & Parker, 2020; Wang et al., 2017). In particular, avoidance job crafting activities such as reducing hindering job demands have been found to be associated with both negative intrapersonal (Petrou et al., 2012; Tims et al., 2013b, 2015) and interpersonal outcomes (Fong et al., 2021; Tims et al., 2015). However, we argue that both the avoidance and approach dimensions of job crafting may result in negative interpersonal outcomes, as job crafting in general involves the mobilization of resources in the work unit and may lead to a potential loss of resources among others. For example, coworkers who carry out approach job crafting by seeking the support or occupying the time of leaders limit other employees’ access to these resources (Tims et al., 2015).

Second, studies of job crafting have generally used person–job fit theory (Kooij et al., 2017), conservation of resources theory (Harju et al., 2016), or social information theory (Tims & Parker, 2020) to examine how job crafting activities, as conceptualized in job demands–resources theory, influence the work attitudes and behaviors of job crafters. These theoretical frameworks have been used mainly to understand the positive effects of job crafting. In contrast, we used relative deprivation theory to explain how job crafting may have a negative interpersonal effect. By revealing the mediating role of relative deprivation and the moderating role of a zero-sum mindset in the relationships between coworkers’ job crafting and employees’ prosocial behaviors and social undermining, our study provides an additional theoretical explanation for the effects of job crafting. Our study also extends the application of relative deprivation theory to job crafting and proactivity research.

Finally, we contribute to the broader literature on proactive behavior and (un)ethical behavior. Although proactive behavior may seem beneficial, researchers have recently shown interest in its negative effects, such as decreased individual performance and team learning or increased employee stress (for a review, see Parker et al., 2019). In this research, we examined how proactive employee behavior, such as job crafting, is associated with fewer prosocial reactions and more social undermining. We thus contribute to proactivity research by enriching the understanding of the interpersonal consequences of proactive behavior. As to the antecedents of unethical behavior, researchers have mainly focused on individual disposition, leadership, or the team context (for a review, see Kish-Gephart et al., 2010). We went beyond previous research by identifying a new antecedent of unethical behavior (i.e., job crafting). Our research also opens the discussion on the ethical issues of individual job crafting while working in a team.

Theory and Hypotheses

Job Crafting and Its Interpersonal Outcomes

Tims et al. (2012) employ the job demands–resources theory to define job crafting as the various strategies or activities that employees initiate to better balance their job demands and resources with their personal needs and abilities. Job crafting activities include increasing social resources (e.g., leader coaching, feedback, and social support), structural resources (e.g., job variety, autonomy, and opportunities for personal development), and challenging job demands (e.g., new projects and additional tasks), as well as decreasing hindering job demands (e.g., the cognitive demands of jobs; Tims et al., 2012). Research has attempted to examine the effects of job crafting by focusing on the interdependence of work contexts. Tims et al. (2015) found that employees experience a heavier workload and more conflict when their coworkers decrease their hindering job demands (e.g., by reducing their working time). Fong and colleagues (2021) meanwhile found that supervisors decrease their social support for avoidance job crafters.

Using relative deprivation theory as the theoretical framework, we propose that coworkers’ job crafting causes employees to develop feelings of relative deprivation and react negatively to job crafters. We adopted relative deprivation theory for the following reasons. First, relative deprivation theory explicitly argues that feelings of deprivation stem from social comparison with referent others. This basic tenet is consistent with our goal to examine the effects of coworkers’ job crafting on employees’ behaviors in a team context. Second, the nature of job crafting and its potential interpersonal consequences can be conceptualized in terms of relative deprivation. Job crafting is regarded as a way to seek social or structural resources (Harju et al., 2016; Tims et al., 2015). Given that the resources within teams or organizations are limited, coworkers’ acquisition of resources is likely to lead to other employees’ feelings of relative deprivation. Third, relative deprivation theory further argues that feelings of relative deprivation and subsequent behaviors vary according to individual differences. These features of relative deprivation theory helped us address our research question (i.e., how do employees react to coworkers’ job crafting?) and build a theoretical model in which coworkers’ job crafting has an indirect effect on employees’ behaviors through relative deprivation, a psychological process moderated by individuals’ zero-sum mindset.

Although “relative deprivation can be interpreted as falling within the purview of social comparison theory” (Olson & Hazlewood, 1986, p. 10), we adopted relative deprivation theory rather than general social comparison theory due to our focus on the negative interpersonal consequences of job crafting. Social comparison theory (Festinger, 1954) suggests that individuals tend to self-evaluate by comparing themselves with similar others. Later, this notion was expanded to include emotions (e.g., fear; Schachter, 1959), self-enhancement (i.e., restoring one’s self-esteem by comparing oneself with others worse off; Wills, 1981), and self-improvement (i.e., seeking a positive example of the domain under evaluation; Wilson & Benner, 1971). According to social comparison theory, reactions to social comparison can differ for upward and downward comparisons. For example, when employees realize that they may use a similar approach (e.g., job crafting) and be in the same favorable position as others (upward comparison), they may develop positive feelings, such as feelings of self-enhancement. However, according to relative deprivation theory, deprivation is described as “relative,” because it is a feeling that is not necessarily “objectively the most destitute” (Martin et al., 1984, p. 485). Therefore, it is often true that those who are the least deprived in an objective sense are the ones most likely to experience deprivation. We propose that an increase in relative deprivation is fostered in response to both upward and downward social comparisons of coworker job crafting (Buunk et al., 2003).

Coworkers’ Job Crafting and Employees’ Feelings of Relative Deprivation

According to relative deprivation theory (Crosby, 1984; Mark & Folger, 1984), when employees believe that the desired or deserved resources have been taken away by others, they tend to feel relatively deprived and respond negatively to coworkers and organizations. The sense of deprivation is relative and arises from comparisons with referents, such as coworkers. Therefore, an important precondition of feeling deprived is the presence of others who seem to be more advantaged (Hu et al., 2015). Given the finite nature of resources in an organization (Epitropaki et al., 2016), job crafters may be viewed by focal employees as competitors for these resources, and may thus become targets of employees’ social comparison.

Coworkers’ job crafting makes focal employees feel that they have been deprived of desired or deserved resources, regardless of the focal employees’ level of job crafting or direction of comparison. Job crafting is a way to increase one’s resources at work (Tims et al., 2012, 2015). Job crafters might gain the resources that the focal employees would otherwise have. Viewing others conducting job crafting activities is likely to trigger a social comparison process wherein the focal employees may perceive their own situation are getting worse because of the job crafters. According to relative deprivation theory, such social comparison may result in feelings of relative deprivation (Crosby, 1984). Given that the total resources are finite and limited in an organization (Epitropaki et al., 2016; e.g., career development opportunities and challenging work assignments), focal employees may perceive job crafters as gaining resources at the expense of others, or as gaining resources that focal employees are entitled to (Bolino et al., 2010; Tims & Parker, 2020; Tims et al., 2015). Even if focal employees conduct a higher level of job crafting than their coworkers, coworkers’ job crafting may cause the focal employees to feel deprived of desired or deserved resources. For example, when coworkers seize opportunities for career development or participation in challenging tasks, such opportunities become unavailable for other employees (Tims et al., 2015). Employees may therefore feel deprived of resources that they could otherwise acquire. Meanwhile, the demand reduction activities of coworkers make it likely that other employees will have to take over the handling of these job demands (Tims & Parker, 2020; Tims et al., 2015). Employees may have to deplete their personal resources to deal with these additional job demands, and therefore feel relatively deprived. Research has also demonstrated that coworkers’ reduction of hindering job demands leads to increased workload and job strain for employees (Tims et al., 2015). Under such conditions, employees will experience relative deprivation, as from their perspective, their coworkers are evidently more advantaged (Schaubroeck & Lam, 2004). Therefore, we propose a positive relationship between coworkers’ job crafting and employees’ feelings of relative deprivation.

Hypothesis 1 (H1)

Coworkers’ job crafting is positively related to employees’ feelings of relative deprivation.

Employees’ Feelings of Relative Deprivation, Prosocial Behavior, and Social Undermining

According to relative deprivation theory (Crosby, 1984), the feelings of relative deprivation that stem from comparisons with referent others affect employees’ interpersonal behaviors. In this study, we examined two types of interpersonal behaviors: employee prosocial behavior and social undermining.

Prosocial behavior refers to various actions that are taken to benefit others, such as generosity, cooperation, and reciprocity of support (Penner et al., 2005; Taylor & Curtis, 2018; Thielmann et al., 2020). We propose a negative relationship between employees’ feelings of relative deprivation and prosocial behaviors for the following reasons. First, employees who experience relative deprivation may wish to compensate for their perceived disadvantages in resource mobilization relative to job crafters (Crosby, 1984; Sun et al., 2020). Therefore, they may reduce their prosocial behaviors toward job crafters, as such behaviors often incur personal resources to benefit job crafters (Sun et al., 2020; Thielmann et al., 2020). This argument is in line with conservation of resources theory (Hobfoll, 1989), which states that individuals are motivated to protect themselves from actual or potential resource loss. Second, the focal employees who perceive themselves as relatively deprived may experience resentment and anger toward job crafters (Crosby, 1984; Zoogah, 2010). They may see themselves as the victims of unfair treatment resulting from the actions of job crafters (Crosby, 1984). To relieve these negative emotions, these employees are likely to decrease their prosocial helping behavior toward job crafters (Zhang et al., 2016). Studies have also empirically shown that destructive emotional states such as anger tend to inhibit help-giving (Zhang et al., 2016) and induce aggressive behaviors (Lemay et al., 2012).

Hypothesis 2a (H2a)

Employees’ feelings of relative deprivation are negatively related to employee prosocial behavior.

Social undermining is defined as “behavior intended to hinder, over time, the ability to establish and maintain positive interpersonal relationships, work-related success, and favorable reputation” (Duffy et al., 2002, p. 332). Such behavior includes spreading rumors about a coworker, intentionally delaying work to slow down a coworker, or giving a coworker misleading information (Duffy et al., 2002; Quade et al., 2019). When employees experience relative deprivation, they may attempt to harm the coworkers whom they perceive to be better off to maintain a positive self-image or status and reduce their sense of relative deprivation (Sun et al., 2020). Social undermining, which is covert and insidious (Duffy et al., 2002; Reh et al., 2018), is considered to be a particularly attractive strategy for enhancing oneself at the expense of others. Furthermore, feelings of relative deprivation are accompanied by negative emotions such as anger or envy (Crosby, 1984; Zoogah, 2010). Research has shown that negative attitudes toward coworkers often lead to harmful interpersonal activities such as social undermining (Reh et al., 2018; Sun et al., 2020). Therefore, we suggest that employees who experience relative deprivation undermine their coworkers.

Hypothesis 2b (H2b)

Employees’ feelings of relative deprivation are positively related to social undermining.

Combining Hypotheses 1, 2a, and 2b, we suggest that employees’ feelings of relative deprivation play a mediating role in the relationships between coworkers’ job crafting and employee prosocial behavior, and between coworkers’ job crafting and employee social undermining. Specifically, we argue that when confronted with job crafters, employees are likely to feel deprived of the resources that they desire. This sense of relative deprivation leads to employees carrying out fewer prosocial behaviors and more social undermining to conserve their remaining resources and maintain their positive image. Consistent with our arguments, Van Dyne and Ellis (2004) argued that coworkers who are seen by other employees as proactively doing too much could become victims of retaliatory activities such as attempts to harm their reputations. A recent study also showed that proactive personalities in employees trigger coworker envy and lead to fewer prosocial behaviors and more social undermining, thus indirectly supporting our hypothesis (Sun et al., 2020).

Hypothesis 3a (H3a)

Employees’ feelings of relative deprivation mediate the indirect relationship between coworkers’ job crafting and employee prosocial behavior.

Hypothesis 3b (H3b)

Employees’ feelings of relative deprivation mediate the indirect relationship between coworkers’ job crafting and employee social undermining.

The Moderating Effect of Employees’ Zero-Sum Mindset

Relative deprivation theory suggests that the strength of the effect of comparisons on relative deprivation depends on employees’ attitudes toward the resources (Buunk et al., 2003; Crosby, 1984). We propose that the effect of coworkers’ job crafting on employees’ relative deprivation varies with employees’ zero-sum mindset.

Employees with a high zero-sum mindset assume there is a finite amount of resources and compete with team members for these resources. They believe that one person’s resource gain is likely to be at the expense of others (Sirola & Pitesa, 2017). Therefore, these employees are more likely to regard coworkers’ job crafting as depriving them of desired resources, thus leading to stronger feelings of relative deprivation. Furthermore, employees tend to make attributions about the behaviors of others when they encounter events relative to themselves or that deviate from the normal routine, especially when these behaviors affect their desired outcomes or goals (Tims & Parker, 2020). As job crafting is not conducted in a completely isolated environment and is regarded as deviating from activities normally carried out under the existing management system (Dierdorff & Jensen, 2018), employees may unconsciously make assumptions about the job crafters’ motives (i.e., prosocial or proself; Tims & Parker, 2020). Because employees with a high zero-sum mindset assume situations to be win–lose and lack interpersonal trust (Davidai & Ongis, 2019; Sirola & Pitesa, 2017), they are more likely to attribute personal motives to coworkers engaging in job crafting. They may regard such coworkers as selfish and as engaging in job crafting at the expense of others (De Dreu, 2007). For example, they may believe that the workflow changes made by coworkers hinder their own processes and disrupt the regular workflow on which they rely to finish tasks. This belief may strengthen their feelings of relative deprivation.

In contrast, employees with a low zero-sum mindset do not regard the amount of resources as fixed (Sirola & Pitesa, 2017). They are likely to believe that coworkers’ job crafting can generate additional resources, and that improvements in the work conditions of coworkers do not necessarily have to be realized at their expense. Such employees are not likely to feel deprived. Moreover, they believe in the existence of win–win situations, and tend to make prosocial motive attributions (Sirola & Pitesa, 2017; Tims & Parker, 2020). They believe that coworkers’ job crafting can benefit not only those coworkers but also other team members (De Dreu, 2007). For example, they may believe that coworkers can add new and important tasks to the team workload by increasing challenging demands, and improve work methods that are shared with other employees (Tims & Parker, 2020). They may also believe that coworkers’ job crafting can promote information sharing, team creativity, and job performance (Tims et al., 2013b). The likelihood of these employees experiencing feelings of relative deprivation is therefore low.

Hypothesis 4 (H4)

Employees’ zero-sum mindset moderates the relationship between coworkers’ job crafting and employees’ feelings of relative deprivation, such that this positive relationship is stronger for employees with a high rather than low zero-sum mindset.

Combining Hypotheses 4, 2a, and 2b, we propose a moderated mediation model in which employees with a high zero-sum mindset engage in fewer prosocial behaviors and more social undermining, due to their higher likelihood of feeling relatively deprived. In contrast, employees with a low zero-sum mindset are less likely to feel relatively deprived in response to coworkers’ job crafting, and thus engage in more prosocial behavior and avoid social undermining.

Hypothesis 5a (H5a)

Employees’ zero-sum mindset moderates the indirect relationship between coworkers’ job crafting and employees’ prosocial behavior via employees’ feelings of relative deprivation, such that the indirect relationship is stronger for employees with a high rather than low zero-sum mindset.

Hypothesis 5b (H5b)

Employees’ zero-sum mindset moderates the indirect relationship between coworkers’ job crafting and employees’ social undermining via employees’ feelings of relative deprivation, such that the indirect relationship is stronger for employees with a high rather than low zero-sum mindset.

Method

Sample and Procedures

We collected data from a large steel corporation in southern China. This corporation has enjoyed a high reputation in the steel industry for more than 60 years and maintains branches in different parts of China. We conducted our research with the leaders and employees working at the company headquarters. Our study involved employees from departments such as marketing (19.83%), science and technology quality (15.15%), strategic operation (13.22%), company administration [including human resources (HR), 12.39%], research (11.29%), and finance (9.64%). The company encourages employees to take initiative and propose new ideas to improve work procedures. Employees can therefore conduct job crafting activities, such as increasing structural job resources, to adapt their job requirements to the dynamic environment. These employees also have to collaborate with each other to achieve team goals. To ensure the interdependent nature of work teams in our sample, we measured team task interdependence (Cronbach’s alpha = 0.86) with the three-item scale developed by Liden et al. (1997). The results showed that, on average, the teams had high task interdependence (M = 5.74, SD = 1.00).

All of the participants were informed of the voluntary nature of their participation, the procedures for questionnaire completion, and the confidentiality of their responses. Prior to data collection, we obtained a name roster from the company’s HR manager, which contained a description of the team structure (i.e., who was the leader and who were the employees). Using the name roster, we assigned codes to the participants (i.e., the leader in team A was coded as A01, and the employees were coded as A0101, A0102, and A0103) to facilitate identification of leader–employee dyads and of matching participant responses at two time points. At Time 1, we distributed questionnaires to 387 full-time employees who volunteered to participate in our survey. They self-reported their job crafting behaviors and zero-sum mindset and provided their demographic information. Approximately 319 employees completed the questionnaires, with a response rate of 82.43%. At Time 2 (1 month later),Footnote 1 we distributed another set of questionnaires to these 319 employees and their 86 team leaders. The employees rated their feelings of relative deprivation and social undermining. The leaders evaluated their employees’ prosocial behavior and provided their own demographic information.

Our final sample consisted of 313 employees and their 85 team leaders who returned completed questionnaires, with a response rate of 98.84% for employees and 84.88% for team leaders. The average age of the employees was 41.10 years (SD = 10.18). Most of them were men (65.81%) and had obtained a bachelor’s degree or higher (85.94%). The average age of the team leaders was 46.59 years (SD = 7.08). Similarly, most of them were men (86.30%) and had obtained a bachelor’s degree or higher (98.63%).

Measures

As all of the scales had originally been developed in English, we translated them into Chinese according to the translation–back–translation procedure developed by Brislin (1986). We measured all of the constructs using a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree) unless otherwise indicated.

Coworkers’ Job Crafting

Employees self-reported their job crafting on the 21-item scale developed by Tims et al. (2012). A sample item is “I try to ensure that my work is emotionally less intense” (1 = never, 7 = always). Cronbach’s alpha was 0.91. This scale has been commonly used in job crafting research (e.g., Bakker & Oerlemans, 2019; Dierdorff & Jensen, 2018). Because our primary goal was to examine the potential costs of job crafting in general, we used an overall composite score to test our hypotheses (Dierdorff & Jensen, 2018). The results from a multilevel confirmatory factor analysis (MCFA) show that the second-order factor model provided an adequate fit to the data (χ2 = 129.07, df = 50, p < 0.001, CFI = 0.95, TLI = 0.93, RMSEA = 0.07, SRMR-within = 0.07, SRMR-between = 0.00, AIC = 8,445.45, BIC = 8,595. 30).Footnote 2 Following previous research, we calculated the composite score for coworkers’ job crafting by averaging the self-ratings of all employees for the 21 items, excluding the focal employee’s score. Therefore, for each focal employee, there was a corresponding average score for coworkers’ job crafting (Hu et al., 2015).

Employees’ Feelings of Relative Deprivation

Employees reported their feelings of relative deprivation using the five-item scale developed by Callan et al. (2011). A sample item is “I feel deprived when I compare the job I have to the one other people who have similar qualifications (education, experience, skills).” Cronbach’s alpha was 0.84.

Employees’ Zero-Sum Mindset

Employees provided ratings of their zero-sum mindset on a six-item “zero-sum construal of success” scale adapted from Sirola and Pitesa (2017). The original scale was developed to measure a person’s view that success for some implies a loss for others. We slightly adapted the scale to evaluate employees’ belief that resource gain for some employees comes at the expense of others. A sample item is “More good jobs for some employees means fewer good jobs for other employees.” Cronbach’s alpha was 0.82.

Employee Prosocial Behavior

Team leaders rated their employees’ prosocial behavior using the three items developed by De Dreu and Nauta (2009). A sample item is “This employee helps colleagues to solve work-related problems.” Cronbach’s alpha was 0.75.

Employee Social Undermining

Employees self-reported their social undermining using the 13-item scale from Duffy et al. (2006). A sample item is “I sometimes give my coworkers the silent treatment.” Cronbach’s alpha was 0.81.

Control Variables

We controlled for employees’ gender (0 = male, 1 = female), age, and education (1 = high school and below, 2 = three-year college degree, 3 = bachelor’s degree, 4 = master’s degree or higher), because researchers have suggested that these demographic variables are associated with social interactions, social comparisons, and anti/prosocial behaviors (Duffy et al., 2006; Reh et al., 2018). For example, compared with men, women are more likely to craft their jobs by increasing social interactions with others (Dierdorff & Jensen, 2018). Meanwhile, older or more educated employees tend to have higher status and seniority and are therefore more likely to be the targets of social comparison (Campbell et al., 2017). Moreover, as our study focused on the interpersonal effects of job crafting, i.e., employees’ psychological and behavioral reactions to coworkers’ job crafting, we controlled for the focal employees’ job crafting to eliminate its confounding influence.

Analytical Strategy

Given the nested nature of our data (i.e., employees were nested within teams; ICC(1) for prosocial behavior was 0.44, F = 4.11, p < 0.001, while ICC(1) for social undermining behavior was 0.17, F = 1.77, p < 0.001), we conducted multilevel analyses of our data using Mplus 7.4. Our primary goal was to model variances at the individual level. Thus, following Anand et al. (2010), we used level-2 fixed effects for all of the coefficients and specified random effects for the intercept, which allowed us to examine between-group variances for the means of employees’ feelings of relative deprivation, prosocial behavior, and social undermining. Following Hofmann and Gavin (1998), we grand-mean-centered all of the predictors, and created the interaction term by multiplying the centered variables of coworkers’ job crafting and employees’ zero-sum mindset (Anand et al., 2010; Hu et al., 2015).

To estimate the indirect (mediation) and conditional indirect effects (moderated mediation), we had to calculate the compound coefficients, which were not normally distributed and may produce Type I errors (Shrout & Bolger, 2002). We used Monte Carlo simulation procedures (20,000 repetitions) with R to obtain bias-corrected 95% confidence intervals (CIs; Preacher et al., 2010). We chose this method because it does not assume that the interaction terms (indirect effects and moderation effects) are “normally distributed and yields asymmetric CIs that are faithful to the skewed sampling distributions of the product term” (Preacher et al., 2010, p. 223).

Results

Multilevel Confirmatory Factor Analysis (MCFA)

We used MCFA procedures with Mplus 7.4 to assess the distinctiveness of our five measures via maximum likelihood estimation. Given the small size of our sample, we simplified the measurement models by sequentially averaging the items to create three indicators for each construct (Bagozzi & Edwards, 1998).

The results suggested that the data had a better fit with the baseline five-factor model (χ2 = 85.31, df = 80, RMSEA = 0.02, CFI = 1.00, TLI = 1.00, SRMR-within = 0.03, SRMR-between = 0.00) than with alternative models (see Table 1). As the five-factor model had CFI and TLI values greater than 0.90, RMSEA less than 0.06, SRMR-within less than 0.08 (Hu & Bentler, 1999), and SRMR-between less than 0.14 (Hsu et al., 2015), it provided an adequate fit to our data. Therefore, we retained all five constructs in the subsequent analyses.Footnote 3

Table 1 Multilevel confirmatory factor analysis
Full size table

Table 2 presents the means, standard deviations, correlations, and reliability of the variables used in this study. In line with our expectations, feelings of relative deprivation had a negative correlation with prosocial behavior (r =  − 0.22, p < 0.01) and a positive correlation with social undermining (r = 0.20, p < 0.01).

Table 2 Means, standard deviations, reliabilities, and correlations among the study variables
Full size table

Hypothesis Testing

Table 3 (Model 2) shows that coworkers’ job crafting was positively related to feelings of relative deprivation (B = 0.43, SE = 0.19, p < 0.05). Table 4 (Model 3) shows that feelings of relative deprivation were negatively related to prosocial behavior (B =  − 0.14, SE = 0.06, p < 0.05). Table 5 (Model 3) shows that feelings of relative deprivation were positively related to social undermining (B = 0.05, SE = 0.01, p < 0.01). Thus, Hypotheses 1, 2a, and 2b were supported (Fig. 1).

Table 3 Regression on employees’ feelings of relative deprivation
Full size table
Table 4 Regression on employee prosocial behavior
Full size table
Table 5 Regression on employee social undermining
Full size table
Fig. 1
figure 1

The overall hypothesized model. “ + ” means that the proposed effect is positive; “–” means that the proposed effect is negative

Full size image

We used the Monte Carlo simulation procedures recommended by Preacher et al. (2010) to calculate the indirect effects of coworkers’ job crafting on prosocial behavior via feelings of relative deprivation. The results showed that coworkers’ job crafting was negatively related to prosocial behavior via feelings of relative deprivation [estimator =  − 0.06, 95% CI (− 0.17, − 0.00) excluding 0]. We followed similar procedures to examine the mediating role of feelings of relative deprivation in the relationship between coworkers’ job crafting and employees’ social undermining. The results showed that coworkers’ job crafting was positively related to social undermining via feelings of relative deprivation [estimator = 0.02, 95% CI (0.00, 0.05) excluding 0]. Thus, Hypotheses 3a and 3b were supported.

To test Hypothesis 4, we included in our model the main effects of coworkers’ job crafting and employees’ zero-sum mindset, and an interaction effect formed by multiplying coworkers’ job crafting with employees’ zero-sum mindset (as shown in Model 4 of Table 3). The results indicated a significant coefficient for the interaction between coworkers’ job crafting and employees’ zero-sum mindset (B = 0.28, SE = 0.05, p < 0.05). We further plotted the interaction effect (see Fig. 2) and conducted simple slope tests (Aiken & West, 1991). The results indicated that the relationship between coworkers’ job crafting and employees’ zero-sum mindset was nonsignificant for individuals with a low zero-sum mindset (− 1 SD; simple slope =  − 0.03, SE =  − 0.16, n.s.), but this positive effect was strengthened to a significant level for individuals with a high zero-sum mindset (+ 1 SD; simple slope = 0.73, SE = 2.53, p < 0.01). Thus, Hypothesis 4 was supported.

Fig. 2
figure 2

Relationship between coworkers’ job crafting and employees’ feelings of relative deprivation at high and low levels of employees’ zero-sum mindset

Full size image

To test our full model, we used the Monte Carlo simulation procedures recommended by Preacher et al. (2010) to calculate the conditional indirect effects of coworkers’ job crafting on prosocial behavior via feelings of relative deprivation for each level of zero-sum mindset. The results revealed that coworkers’ job crafting had a negative effect on prosocial behavior via feelings of relative deprivation for individuals with a high zero-sum mindset [+ 1 SD; indirect effect =  − 0.09; 95% CI (− 0.24, − 0.00) excluding 0], and a nonsignificant effect for individuals with a low zero-sum mindset [− 1 SD; indirect effect = 0.00; 95% CI (− 0.06, 0.06) including 0]. Meanwhile, we found a significant difference between the indirect effects at high and low levels of zero-sum mindset [diff =  − 0.09, 95% CI (− 0.25, − 0.00) excluding 0]. Using the same simulation procedures, we found that coworkers’ job crafting had a positive effect on social undermining behavior via feelings of relative deprivation for individuals with a high zero-sum mindset [+ 1 SD; indirect effect = 0.04; 95% CI (0.01, 0.08) excluding 0], and a nonsignificant effect for individuals with a low zero-sum mindset [− 1 SD; indirect effect =  − 0.00; 95% CI (− 0.03, 0.02) including 0]. Meanwhile, we found a significant difference between the indirect effects at high and low levels of zero-sum mindset [diff = 0.04, 95% CI (0.00, 0.09) excluding 0]. Therefore, Hypotheses 5b and 5b were supported.

Supplementary Analysis

To check the robustness of our results, we conducted a series of supplementary analyses. First, we followed the recommendations of Bernerth and Aguinis (2016) to examine whether the hypotheses were still supported after removing the control variables (i.e., employees’ gender, age, education, and job crafting). The results still supported our hypotheses.

Second, we examined whether the different dimensions of job crafting produced consistent results. When only one dimension (approach or avoidance) was included in the regression, the results were highly consistent (approach crafting main effect: B = 0.20, SE = 0.10, p = 0.052, interaction effect: B = 0.19, SE = 0.08, p < 0.05; avoidance crafting main effect: B = 0.26, SE = 0.08, p < 0.01, interaction effect: B = 0.15, SE = 0.06, p < 0.05). Yet, when we included both approach and avoidance crafting and their interactions with employees’ zero-sum mindset, only the main effect of coworker approach crafting (B = 0.25, SE = 0.12, p < 0.05) and its interaction effect (B = 0.25, SE = 0.05, p < 0.01) were significant. The main effect of coworker avoidance crafting (B =  − 0.04, SE = 0.14, n.s.) and its interaction effect (B = 0.05, SE = 0.09, n.s.) became nonsignificant. The reason for this result may be that there was a strong correlation between coworkers’ approach crafting and avoidance crafting (r = 0.77, p < 0.001). When both factors were included in the regression, they may have diminished each other’s effect (e.g., Dierdorff & Jensen, 2018; Hong et al., 2016). The strong correlation between the two types of job crafting also supported our use of an overall score for job crafting in our regression analyses.

Finally, we tested whether self-crafting interacted with coworker crafting. The results showed that the interaction effect was not significant for feelings of relative deprivation (B = 0.26, SE = 0.20, n.s.), social undermining (B =  − 0.27, SE = 0.17, n.s.), or prosocial behavior (B = 0.00, SE = 0.06, n.s.).

Discussion

Using two-wave data from a sample of 313 employees and their supervisors from 85 teams, we found that coworkers’ job crafting induced employees’ feelings of relative deprivation. Such feelings, in turn, decreased employees’ prosocial behavior and increased their social undermining. Additionally, employees’ zero-sum mindset influenced the level of relative deprivation stemming from coworkers’ job crafting. Specifically, employees with a high rather than low zero-sum mindset were more likely to feel relatively deprived and thus engaged in fewer prosocial behaviors and more social undermining.

We focused on total job crafting instead of its separate dimensions because we assumed that both the approach and avoidance job crafting of coworkers would increase employees’ feelings of relative deprivation and result in negative interpersonal outcomes. The results of our supplementary analyses confirmed our assumption by showing that both approach and avoidance job crafting, as well as their interactions with employees’ zero-sum mindset, had a significant relationship with perceived relative deprivation. In addition, all of our hypotheses remained valid after removing the control variables. Therefore, our findings were confirmed to be robust.

Theoretical Implications

First, the present study contributes to the job crafting literature by taking into account the interdependent work context and demonstrating the negative effects of coworkers’ job crafting on employees’ psychological and behavioral reactions. Research on job crafting has focused on the positive and intrapersonal effects of job crafting (e.g., Bakker & Oerlemans, 2019; Bakker et al., 2012; Petrou et al., 2012), but neglected the potential costs of job crafting. In this study, we found that coworkers’ job crafting could induce employees to engage in fewer prosocial behaviors and more social undermining by evoking feelings of relative deprivation. This finding supports the notion that negative attitudes are derived at least partially from comparisons with seemingly more advantaged coworkers (Buunk et al., 2003; Hu et al., 2015; Schaubroeck & Lam, 2004; Sun et al., 2020). In addition, researchers have suggested that coworkers’ job crafting may not be a proximal but distal antecedent of employees’ behaviors (Tims & Parker, 2020; Tims et al., 2015). Our findings support this view, as they suggest that the association between coworkers’ job crafting and employees’ behaviors operates indirectly through a psychological mechanism, and that such an association depends on individual factors.

Second, by using relative deprivation theory (Crosby, 1984; Mark & Folger, 1984) as our theoretical framework, we expand the applications of relative deprivation theory. The literature on relative deprivation in organizations has mainly focused on comparing outcomes such as salaries, benefits, or opportunities for career advancement (Erdogan et al., 2018). This has prompted many scholars to recommend extending the applications of relative deprivation theory (e.g., Buunk et al., 2003; Hu et al., 2015; Sun et al., 2020). Our study does so by examining the role of social comparison with job crafting coworkers in fostering feelings of relative deprivation.

Third, our research sheds light on the important boundary conditions of the interpersonal effects of job crafting. Although researchers have attempted to examine the negative effects of job crafting on employees, they have largely ignored the boundary conditions (Tims et al., 2015). Meanwhile, research on the general positive effects of job crafting has mainly focused on the moderating role of work contexts or cultural differences, and neglected the potential moderating role of individual differences (Zhang & Parker, 2019). Our study advances this line of research by showing that employees’ zero-sum mindset serves as a necessary moderator of how employees’ relative deprivation stems from coworkers’ job crafting. This finding is in accordance with previous research demonstrating the strong influence of individual differences, such as social comparison orientation (Buunk et al., 2003) and prosocial motivation (Sun et al., 2020), on employees’ reactions in the comparison process. Additionally, our results contribute to knowledge of the boundary conditions of relative deprivation theory (Crosby, 1984; Mark & Folger, 1984). Although relative deprivation theory points out that individual characteristics can lead to different levels of relative deprivation in the comparison process, empirical research directly examining the moderating role of individual differences is limited. By investigating the moderating effect of employees’ zero-sum mindset, the present study increases our understanding of the contexts under which relative deprivation theory works.

Finally, our findings contribute to the literature on proactive employee behavior and (un)ethical behavior. Our study revealed that proactive employee behavior, such as job crafting, could indirectly lead to fewer prosocial behaviors and more social undermining. Although there is considerable evidence of the positive effects of various forms of proactivity on outcomes such as work performance and career success, the body of evidence suggesting that proactivity is not always positive is growing (Parker et al., 2019). Recent studies have focused on the negative effects of proactive behavior such as on employee creativity (Breidenthal et al., 2020) and organizational citizenship behavior (Koopman et al., 2016). Likewise, our study helps to develop a more comprehensive nomological network of proactive behavior. Additionally, our results shed light on new precursors of (un)ethical employee behaviors. Research has examined the antecedents of (un)ethical employee behaviors such as individual characteristics, moral issues, organizational climate, and leadership (e.g., Finegan, 1994; Ho et al., 2015; Mai et al., 2021; Mayer et al., 2010). Our study adds to this line of research by putting forward job crafting as another antecedent.

Practical Implications

This study has implications for management practices. First, in contrast to previous research, which has largely examined the positive effects of job crafting, our study calls attention to its detrimental consequences. Organizations and managers should be aware that coworkers’ job crafting can lead employees to engage in fewer prosocial behaviors and more social undermining, and therefore increase interpersonal conflict within teams. HR departments should therefore organize team-building activities frequently to enhance team cohesion and reduce potential interpersonal conflicts. Second, given the role of employees’ feelings of relative deprivation in creating negative interpersonal outcomes, organizations should take measures to reduce such feelings. For example, organizations could investigate or monitor the underlying motives of job crafting (Dierdorff & Jensen, 2018) and encourage more prosocial job crafting to benefit the team as a whole. They should also ensure the procedural justice of resource allocation to attenuate employees’ feelings of relative deprivation (Kim et al., 2018). Finally, organizations could consider employees’ zero-sum mindset when recruiting new employees. As a high zero-sum mindset exacerbates the negative effects of coworkers’ job crafting, managers could also communicate with employees to weaken their belief in the limited nature of resources. Organizations could encourage employees to generate more resources but refrain from competing for them.

Limitations and Future Directions

Despite the theoretical and practical implications of our study, there are limitations that require further investigation. First, we focused only on one type of proactive behavior, i.e., that which is self-starting, future-focused, and change-oriented (Parker et al., 2019). However, individuals can engage in proactive behavior via different methods, such as speaking up (Morrison, 2011) and creativity (Breidenthal et al., 2020), of which job crafting is only one type (Tims et al., 2012; Wrzesniewski & Dutton, 2001). Although commonalities across multiple forms of proactive behavior can be identified, different aspects are emphasized in different forms. For example, there is more importance attached to navigating the task or strategic context in issue selling relative to other forms of proactive behavior. Meanwhile, resource-based job crafting is more strongly associated with resource reallocation (Parker et al., 2019). Future research could attempt to replicate our findings with different types of proactive behaviors, or examine the interpersonal consequences of other proactive behaviors.

Second, we focused only on the mediating role of relative deprivation and on the moderating role of employees’ zero-sum mindset. Future research could therefore consider other mediators and moderators. While the present study showed that comparison with job crafting coworkers may lead to employees’ feelings of relative deprivation, social comparison theory (Festinger, 1954) suggests that employees can respond differently to comparison. For example, they may envy their coworkers, pursue self-enhancement, or pursue self-improvement, depending on the characteristics of the individuals and situations (Buunk et al., 1990). Therefore, future research could examine the other possible responses of employees to social comparison with job crafters. In addition, while the current research revealed that employees’ zero-sum mindset may exacerbate the negative interpersonal effects of coworkers’ job crafting, it did not consider how cultural orientations (e.g., collectivism vs. individualism) may shape people’s zero-sum mindset (Chen et al., 2002). To examine the generalizability of the moderating role of a zero-sum mindset, researchers could investigate our model in other countries with different cultural backgrounds.

Third, given that our data for feelings of relative deprivation and social undermining were measured from the same source and at the same time, our study could suffer from common method bias. However, the results of our MCFA confirmed the discriminant validity of all of the key variables, which suggests that common method bias had minimal influence. More importantly, employees are the best source for rating employee social undermining, as such behavior is covert and insidious (Duffy et al., 2002; Reh et al., 2018). In addition, despite the time-lagged nature of our data, causal inferences could not be made. Future research should consider using more rigorous longitudinal cross-lagged panel designs or experiments to obtain stronger evidence for the causal relationships among variables.

Conclusion

Wrzesniewski and Dutton (2001) argued that job crafting is neither inherently good nor bad for organizations, but that the nature of its effects depends on the kind of changes job crafters make, as well as their motivation and performance. Other researchers have suggested that as job crafting involves shifts in resources and responsibilities in the work unit (Tims et al., 2012), as well as deviations from established routines (Dierdorff & Jensen, 2018), it may not always be appreciated by coworkers as a morally correct activity (Parker & Collins, 2010). However, most empirical studies have focused on the positive effects of job crafting, thus appearing to support the notion that job crafting is morally right. In contrast, our study found that coworkers’ job crafting could lead to employees’ feelings of relative deprivation, and thus cause employees to engage in fewer prosocial behaviors and more social undermining. Our findings showed that job crafting could generate unintentional negative interpersonal results, especially among those with a high zero-sum mindset. Therefore, our study suggests that employees’ mindsets play a role in their judgment of whether job crafting is morally right or wrong. When employees consider the resources in their work unit to be finite and limited, job crafting is likely to become more of an ethical issue.